The pathogenesis of HIV-1 infection in the central nervous system (CNS) is uncertain. The role of factors which regulate HIV-1 production in the brain or which actually cause damage to neural cells remains to be defined. Using transgenic mouse technology we will address several hypotheses relevant to the neuropathogenesis of HIV-1 in vivo including: 1. Inflammation in the brain activates HIV-1. Using an established line of transgenic mice in which beta-galactosidase (beta-gal) is expressed in the brain under the direction of HIV-1 long terminal repeat (LTR), we will create inflammation with a cerebral stab wound and herpes simplex virus type 1 encephalitis. beta-gal production will be compared in treated and untreated mice. 2. Tumor necrosis factor-alpha (TNF-alpha) is neurotoxic when secreted in the brain. We will generate transgenic mice which express TNF-alpha under the direction of a brain specific promoter, glial fibrillary acidic protein (GFAP), evaluate offspring mice neuropathologically, and compare the neuropathology to that seen in HIV-1 CNS infection. 3. HIV-1, transactivating protein (tat) is neurotoxic when secreted in the brain. We will create two sets of transgenic mice which express tat in the brain under the direction of GFAP and HIV-1 LTR, evaluate the offspring mice neuropathologically, and compare the neuropathology to that seen in HIV-1 CBS infection. 4. TNF-alpha and tat activate HIV-1 LTR in the brain. Transgenic mice expressing TNF-alpha and tat in the brain will be bred with HIV-1 LTR- beta-gal mice, and offspring carrying both transgenes will be assessed for beta-gal production in comparison to littermates who carry either one or neither transgene.